Topic: Appearance of waves in arrangement
- This topic has 8 replies, 3 voices, and was last updated 19 years, 4 months ago by
acousmod.
-
June 26, 2006 at 17:01 #870
acousmodParticipantHi Frits,
This is a very little bug, but it is annoying at least : the accuracy of the drawings of the sounds in the arrangement depends on the vertical zoom value.
It corresponds to the amplitude values only one time on two. When you change the height of a track step by step, it alternates between a “good” and a “strange” drawing.
It is of course more visible when there is a lot of channels or when the height is small.It is not vital, but if you have some time to look at this…
Thanks !
Jean-MarcJuly 10, 2006 at 10:12 #7091
acousmodParticipantSome example :
good :

not good :

good :

not good :

good

etc.July 10, 2006 at 10:17 #7092
ConquistadorParticipantBizarre. I don’t think I have ever seen that.
Are you seeing this in 1.63?
July 10, 2006 at 11:07 #7095
ZynewaveKeymasterThis happens because the rounding errors are different depending on whether the resized wave display is even or odd pixels in height. The ideal solution to this would be to draw anti-aliased waveforms, but this will have a big impact on the graphics update speed. Perhaps within a couple of years a standard computer system will be sufficiently fast so that I can substitute much of my graphics code with true anti-aliased drawing. Another hack solution would be to always restrict channel heights to an odd pixel count, which on the other hand would result in less fluent zooming.
July 10, 2006 at 11:47 #7098
acousmodParticipantAnother hack solution would be to always restrict channel heights to an odd pixel count, which on the other hand would result in less fluent zooming.
For me it would be a perfect solution.
I don’t know how it is made in Nuendo, but the appearance is consistant whatever the zoom value is.
Perhaps that a possibility to have an amplitude zoom slider in the arrange like in the sound editor will also help for small amplitudes.
July 26, 2006 at 15:33 #7165
ConquistadorParticipantAnother hack solution would be to always restrict channel heights to an odd pixel count, which on the other hand would result in less fluent zooming.
Hmmm, I really don’t think I would want less fluent zooming Frits, a step backward there. I get the feeling you do not want this either.
Equally if the impact on the current graphics update speed would be affected in any way I would prefer to defintiely explore another solution.
Unless…you could somehow make the hack (restricting channel heights to an odd pixel count) an optional extra. If you implement it this way then at least users like me can switch this behaviour off.
I would certainly not prefer to see this as a hidden default global setting that cannot be changed. Definitely not IMHO.
If possible Frits if you are to go ahead with this…please implement it as an optional solution.
Thanks.
November 15, 2006 at 18:26 #8159
ZynewaveKeymaster@acousmod wrote:
Another hack solution would be to always restrict channel heights to an odd pixel count, which on the other hand would result in less fluent zooming.
For me it would be a perfect solution.
As part of the ongoing work on the fade-in/out feature, I’ve now changed this so that each channel is always set to an unequal pixel height. Thus you won’t get the alternating thin/fat center line when you zoom vertically. Coming up in 1.73.
November 15, 2006 at 21:18 #8160
ConquistadorParticipantThanks for the update. 😉
November 16, 2006 at 14:46 #8164
acousmodParticipantAs part of the ongoing work on the fade-in/out feature, I’ve now changed this so that each channel is always set to an unequal pixel height. Thus you won’t get the alternating thin/fat center line when you zoom vertically. Coming up in 1.73.
Woah ! Excellent !
Thank you very much Frits !
- You must be logged in to reply to this topic.
